Amazon has officially introduced a series of new AI technologies, notably the Nova Sonic voice model - a step forward to compete with rivals such as Gemini Live of Google or Advanced Voice Mode of OpenAI. Along with that is an update to a video creation technology called Nova Reel.
Nova Sonic is a new voice model from Amazon, with real-time speech processing and the ability to create seamless AI voice for voiceover applications.
According to Amazon, Nova Sonic is built on a consolidated model architecture, meaning all stages such as voice recognition, converting speech to text, creating feedback and converting text to voice are in a single, inseparable system like the old methods.
Notably, Nova Sonic can recognize users' emotions and voice tones, thereby responding naturally and in more depth.
This model is now available on Amazon's Bedrock platform - for developers and can be applied in many fields such as customer service, tourism, education, healthcare, etc.
In addition, Amazon also launched the Nova Reel 1.1 update - AI video creation technology. The new version brings better image quality and processing speed.
Amazon says these improvements will make it easier for developers to create high-quality video content that serves multiple creative or commercial purposes.
